Download Valea LifeScience09 R

IPFocus
LIFE SCIENCES
7TH EDITION
PATENTING PROTEIN PHARMACEUTICALS
IN EUROPE
VALEA
2009
EUROPE: PROTEIN PHARMACEUTICALS
Patenting protein pharmaceuticals
in Europe
Treatment methods based on biotechnology are becoming increasingly important in the
life sciences industries. Ylva Skoglösa and Annika Unge of Valea examine how the EPO
treats protein pharmaceuticals
B
esides the classical medically active substances
(the so-called small molecules), novel treatment
methods based on biotechnology are increasingly
gaining in importance for industry and patient care.
These methods include biopharmaceuticals and biosimilars, therapeutic antibodies, gene and cell therapy,
regenerative medicine, tissue engineering and nanoscale
medicine transporters.
Introducing protein pharmaceuticals
Protein pharmaceuticals form a rapidly growing category in the arsenal of drugs. Many natural proteins are
being used to treat diabetes, anaemia, hepatitis and cancer. What is more, already approved protein pharmaceuticals are being modified to improve their effectiveness. There are now more than 200 approved peptide
and protein pharmaceuticals on the FDA list.
Classic examples of protein and peptide pharmaceuticals are:
• Insulin (diabetes)
• Interferon ß (relapsing MS)
• Interferon γ (granulomatous)
• TPA (heart attack)
• Albumin (hypovolemia/hypoalbuminemia)
• HGH (growth disturbance)
• Factor VIII (haemophilia)
• Erythropoietin (anaemia)
• Calcitonin (enhance bone mass)
• Oxytocin (stimulate labour)
• Vasopressin (antidiuretic)
The rapid growth in the understanding of biological
processes based on the last decade’s research in the
genome and the proteome projects has contributed to a
more thorough understanding of the function of biomedical substances. Modern biotechnology also gives
researchers the possibility to generate and to investigate
broad libraries of substances, greatly simplifying the
identification of candidate substances from naturally
occurring proteins and peptides.
Challenges for the patent system
As for any pharmaceutical substance, obtaining patent
protection for these new protein-based drugs is important. However, in several aspects, the patent system is
largely oriented towards the needs and parameters of
classical small molecule pharmaceuticals, whereas the
requirements for patent protection for biological macromolecules differs from the protection for more easily
definable chemical entities.
Optimally for patent protection, a discovered protein
or peptide is novel: it has never been isolated and disclosed. Then, even if the protein or peptide is a naturally occurring protein or peptide, it is possible to obtain
patent protection for the substance per se, as long as the
protein or peptide has previously not been characterised,
and provided the protein or peptide has an industrial
application. Hence it is not possible to obtain protection
by just listing amino acid sequences without any information about their industrial application. Support for
the technical effect of the protein or peptide, including
that demonstrated by way of examples, is required in the
application. The requirements for proof-of-concept are
steadily increasing at the European Patent Office, more
or less in parallel with the increasing understanding of
biological processes in the field. Laboratory techniques –
the results of which are commonly used to establish
inventive step over the prior art – are already deemed to
have become standard laboratory practice. Inventive
step then becomes a question of whether or not there is
any incentive for the skilled artisan.
Today, new proteins or peptides are often identified via
their amino acid sequences. A novel protein may therefore
be claimed by its amino acid sequence. However, in analogy with T12/81 in the field of chemistry, even if it has not
been possible to sequence the protein or peptide, patent
protection may be obtained by defining the protein or
peptide via physical parameters, such as chromatography
profiles and molecular mass or by the way the protein is
isolated (that is, by product-by-process claims).
L I F E S C I E N C E S I P F O C U S 2 0 0 9 W W W. M A N A G I N G I P. C O M
The oft-preferred identification of a protein or a peptide
via its amino acid sequence, however, leads to the delicate
problem of how to present this sequence to obtain patent
protection that actually corresponds to the invention’s contribution to the art (in line with the teachings of T409/91
and T435/91). For instance, the amino acid sequences of
many naturally occurring proteins are to a large degree
made up of pharmaceutically uninteresting parts, whereas
only a small part of the protein is directly involved in its
actual biological activity, for example forming the binding
pocket of an enzyme or the epitope of an antigen. Those
active parts of a protein often tend to be conserved
between proteins from different species, while other parts
of the protein may be more varied. For example, amelogenins, the major constituents of the enamel matrix of
developing teeth, are a family of proteins that are highly
conserved through vertebrate evolution and demonstrate a
high overall level of sequence homology among all higher
vertebrates examined (80%). Thus, an approximate 4%
difference in amino acid sequence from Homo sapiens
amelogenin to porcine amelogenin does not seem to relate
to its biological activity in either of the animals. Actually,
the human body still recognises the pig’s protein as its own.
Even in the active site of a protein, not all amino
acids may be absolutely necessary for the protein’s biological activity, and some amino acids may easily be
substituted by similar or completely different amino
acids, without the loss of biological activity.
Limiting a patent claim to a specific amino acid
sequence therefore provides very narrow protection for
the protein or peptide; protection which can easily be
circumvented by third parties by the substitution of one
or more of the amino acids which are not absolutely
necessary for biological activity. Rather, the protein or
peptide is preferably identified in a more open manner
as for example “comprising” the amino acid sequence
or as having a certain degree of homology or identity to
the sequence. Interestingly, in T762/07, “identity” was
interpreted as narrower than “homology”.
Inventive step
In classical chemistry, such as T852/91, it is established
that to deny inventive step for novel chemical compounds
because of their structural similarity to known chemical
compounds amounts to an allegation that a skilled person
would have reasonably expected the same or similar usefulness of both the known and the novel compounds as
the means for solving the technical problem underlying
the application in question. Such an expectation would be
justified if the skilled person knew that the existing structural differences of the chemical compounds concerned
were so small that they would have no essential bearing on
those properties, which where important for solving the
technical problem and could be disregarded. In other
words, a chemical entity can be new and inventive even if
the changes in the chemical structure are minimal.
T643/96 ruled that, in the field of classical drug design,
any structural modification of a pharmacologically active
substance is, in the absence of an established correlation
between structural feature and activity, expected a priori
to disturb the pharmacological activity profile of the compound. For the reasons described above, this should not
reasonably be applied directly to protein pharmaceuticals.
However, the EPO unfortunately tends towards granting
patents with a very narrow scope with regard to the
degree of identity required, relying too heavily on case law
concerning classical small molecule chemistry.
What is more, the need for a broad scope of sequence
identity stands in direct contrast to the need for the technical effect to be reproducible across the whole scope of
the claims, for the invention to have inventive step as
ruled in T2/83 and in T939/92. Thus, although a claim
directed to an amino acid sequence being at least 80%
identical to human amelogenin will invariably encompass
proteins that do not possess enamel matrix activity, limiting the claims to amino acid sequences being at least 99%
identical to human amelogenin still can hold such false
positives, depending on exactly which amino acids differ.
On the other hand, a claim limited to a 99% identity will
exclude the porcine amelogenin protein, which is actually proven to promote dental regeneration in humans.
An applicant trying to patent a protein thereby becomes
trapped between trying to obtain reasonably broad protection to hinder competitors from circumventing the patent
all too easily, while still protecting a protein or peptide
which actually has the desired biological activity. A claim
scope that is too broad will almost inevitably lead to a
rejection based on insufficient disclosure of the invention.
A possible solution to this conundrum is to define the
protein through an identity of a reasonably broad percentage with its amino acid sequence in combination with a
further technical effect, such as a desired biological effect,
or a chemical or physical property. In analogy to the ruling
in T723/05, as long as reasonably closed language is used
in a claim, the technical effect is considered to be a limiting
characteristic of a given substance. In T857/06, the denotation of the protein as a “Tumor Necrosis Factor Binding
Protein II” in combination with data on the characteristics
of the protein in the description was important for distinguishing this protein from prior art. Nonetheless, it is
essential for this strategy to include an easy and detailed
test in the description, so that the person skilled in the art,
who all of a sudden is no longer able to execute the new
standard laboratory protocols routinely, can count on a
reasonable expectation of success (see T2/83). In addition,
one needs to be aware of the imperative for the validity of
any priority claim that no amino acid residues in the
sequence as filed are changed, even if inventors wish to
EUROPE: PROTEIN PHARMACEUTICALS
introduce minor and biologically insignificant corrections
to the original listing during the priority year.
Fall-back positions
As patent applications for protein-based drugs are often
filed early in the research process, when it is not yet clear
exactly which amino acid sequence the pharmaceutical
compound of choice will have, it is important to define the
amino acid sequence of the protein or peptide broadly,
while still ensuring that the application recites more specific amino acid sequences of preferred embodiments as
fall-back positions. It is unnecessary to say that a fall-back
on to such severely limited product claims does not in
many cases result in a satisfactory scope of protection for
the inventors. On the other hand, the research is often
focused on trying to identify the biologically active parts
of the protein to be able to produce a smaller molecule
with optimised activity. Such an optimised peptide or protein fragment can then be protected per se in a second generation of patents, even if the protein is already known.
In line with G5/83, even if a protein or peptide is
known per se, if its medical use has previously not been
disclosed, patent protection may be obtained for the
protein or peptide for any medical use. This is in contrast to the normal praxis of the EPO, where it is not
possible to obtain protection for a product for a specific use. In addition, if a protein or peptide has been disclosed for a medical use, but for the treatment of a different pathological condition, it may still be possible to
obtain protection for the protein or peptide in treating
the newly identified pathological condition, that is a dif-
ferent technical effect of the protein or peptide. In addition, it is possible to protect pharmaceutical compositions comprising the protein or peptide, and even antibodies specific to the protein or peptide.
In contrast to peptides, most proteins are large molecules with a delicate, unique and complicated threedimensional structure. The preparation, isolation and
storage of such molecules, while retaining their biological activities, are therefore associated with numerous
problems. Also, typically, natural sources for protein
pharmaceuticals can lead to immune reactions and may
contain viral and pathogen contaminations. Contrary to
the smaller peptide drugs, which often can be synthesised using solid phase chemistry, most protein pharmaceuticals today are produced recombinantly. Thus, the
method of production often constitutes important subject-matter for patent protection, as well as vectors containing the nucleotide sequence encoding the protein or
peptide of interest. Again, due to the rapid development
in modern biotechnology a skilled person is usually considered to be able to clone and express a gene in a fairly straightforward manner and according to T386/94
(for example), with a reasonable expectation of success.
The drafter’s challenge
One of the challenges for the inventor in the field of
protein pharmaceuticals is to draw up a patent application that is detailed enough to provide novelty, inventive
step and sufficiency of disclosure while still providing a
reasonable protection corresponding to the de facto
contribution to the art.
Ylva Skoglösa
Annika Unge
Ylva is a European patent attorney.
She has a PhD in Medical Research
and expertise within the areas of
medical biotechnology, genetics, proteins and substances and devices for
medical or dental treatment.
Ylva regularly assists potential
investors and founding partners in
evaluating third parties’ IP positions, conducting freedom-to-operate and due diligence analyses. Prior to
entering the IP profession, she did research work at the
San Diego State University in California, the Max
Planck Institute in Munich and the BMC in Uppsala.
Before joining Valea in 2002 she worked for three
years as a patent attorney at Plougmann & Vingtoft in
Denmark. Ylva has been working in the IP field since
1999.
Annika is a European patent attorney. She has a PhD in Biochemistry
and specialises in patenting in the
fields of medicine, biochemistry,
microbiology, molecular biology and
gene technology. She works with all
patent issues including pre-patenting
investigations, patent drafting and
prosecution. Annika also works with business-related
IP issues such as patent strategies, questions regarding validity, infringement, due diligence and freedomto-operate issues.
Annika started working in the IP field in 2001.
Before joining Valea in October 2003, she worked as
a patent attorney at another Swedish IP law firm.
Annika also has experience as a lecturer, both for
internal training courses and external courses.
W W W. M A N A G I N G I P. C O M L I F E S C I E N C E S I P F O C U S 2 0 0 9